prd-to-html
Pass
Audited by Gen Agent Trust Hub on Jun 13, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill processes untrusted external content from GitHub issues or local files to generate documentation. While this creates a potential surface for indirect prompt injection, the risk is mitigated by specific mapping instructions.\n
- Ingestion points: Retrieves data from GitHub issues via the
ghCLI or from local file paths specified in the context.\n - Boundary markers: The agent follows a strict mapping of PRD sections to template components, which provides implicit structural constraints.\n
- Capability inventory: The skill has permissions to read local files and write generated HTML output to the
docs/directory.\n - Sanitization: Content is mapped directly into template placeholders; while no explicit sanitization is defined, the target is a static HTML visualization of provided content.\n- [COMMAND_EXECUTION]: Uses the
gh(GitHub CLI) tool to retrieve issue data. This is a standard and appropriate use of project management tools for the purpose of generating documentation (SKILL.md).
Audit Metadata